There are various breast cancer risk prediction models, some of which do a reasonable job in predicting the number of women in a given population with certain characteristics who will be diagnosed with breast cancer in the future. However, these risk models do a poor job in predicting which individual women in the population will be diagnosed. That is, the risk models have poor discriminatory accuracy. We want to improve the discriminatory accuracies of risk models, potentially by incorporating more information about the individuals, for example, information on lifestyle, environmental exposures, and medical record information.
While there are certain behavioral and lifestyle modifications that women can make to help decrease their breast cancer risk, like increasing physical activity, decreasing alcohol intake, and, for very high-risk women, taking chemopreventive medication or undergoing prophylactic surgery, most women do not undergo these modifications. We are also interested in understanding factors related to this decision-making.
To better understand breast cancer risk and prevention, we are analyzing data and biospecimens collected through the Athena Breast Health Network, a cohort of >100,000 women who underwent breast screening at a UC Health clinic since 2011, and the Wisdom Study, an ongoing cohort of >60,000 women undergoing breast screening nationwide.
